Metraipolis › archive › dc_d4382fe12bbc

Frontier vision-language models have overtaken young adults at detecting AI-generated portraits -- but not their calibration — arXiv:2608.30210

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.30210
License (as archived)
CC0-1.0
Archived
2026-10-17T08:32:05.566Z
SHA-256 of the archived text
4ed92f593163de6f9fff7ba01585765fed051696e374b533333f18d18b25b690
Stable URL of this page
https://metraipolis.com/documento/dc_d4382fe12bbc

How to cite

Frontier vision-language models have overtaken young adults at detecting AI-generated portraits -- but not their calibration — arXiv:2608.30210 — arXiv (en), https://arxiv.org/abs/2608.30210 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d4382fe12bbc (Metraipolis archive dc_d4382fe12bbc, sha256 4ed92f593163de6f9fff7ba01585765fed051696e374b533333f18d18b25b690)
BibTeX
@misc{dcd4382fe12bbc,
  title = {Frontier vision-language models have overtaken young adults at detecting AI-generated portraits -- but not their calibration — arXiv:2608.30210},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.30210},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d4382fe12bbc; sha256 4ed92f593163de6f9fff7ba01585765fed051696e374b533333f18d18b25b690},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.30210. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

AI image generators now create face portraits that are hard to tell from real photographs. Vision-language models (VLMs) are increasingly proposed to flag such images. We benchmarked 19 VLMs on the same 198 face portraits -- real photographs and identity-matched ChatGPT-4o and Imagen 3 versions -- under the same task as our earlier study of 1,667 adults (85% correct overall; accuracy fell steeply with age). The June-2026 cohort of 14 models only matched adults in their 20s-30s. Four weeks later the ceiling broke. Among five July-2026 releases under the identical protocol, gpt-5.6-sol reached 92.8% balanced accuracy (five-draw mean 92.1%), clearly above adults in their 20s (88.5%), and claude-fable-5 detected every AI image while averaging 91.9%. Model sensitivity now exceeds young adults decisively (d' up to 3.4 versus ~ 2.4). What has not been overtaken is human calibration. Model criteria spread from c = -1.10 to +1.45 while humans sit near zero at every age; both new leaders are biased (+0.44, -0.97), and only a few mid-ranked models approach the human balance. Changing the labelled examples still flipped about one answer in four. The best machines now out-see young adults here, without matching the human balance between suspicion and trust.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-01 in cs.HC).
Categorie arXiv: Human-Computer Interaction (cs.HC) ; Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2608.30210 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-17T08:13:05.450Z dalla pagina https://arxiv.org/catchup/cs.HC/2026-09-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).